A Research-Based Guide for the Creation and Deployment of a Low-Resource Machine Translation System

Proceedings of Recent Advances in Natural Language Processing, pages 69–79 Varna, Sep 4–6, 2023 https://doi.org/10.26615/978-954-452-092-2_008 69 3D-EX: A Unified Dataset of Definitions and Dictionary Examples Fatemah Almeman∗△ Hadi Sheikhi◦ Luis Espinosa-Anke∗♢ ∗CardiffNLP, School of Computer Science and Informatics, Cardiff University, UK △ College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, KSA ◦ School of Computer Engineering, Iran University of Science and Technology, Iran ♢AMPLYFI, UK {almemanf, espinosa-ankel}@cardiff.ac.uk ha sheikhi@comp.iust.ac.ir Abstract Definitions are a fundamental building block in lexicography, linguistics and computational semantics. In NLP, they have been used for retrofitting word embeddings or augmenting contextual representations in language models. However, lexical resources containing defini- tions exhibit a wide range of properties, which has implications in the behaviour of models trained and evaluated on them. In this pa- per, we introduce 3D-EX, a dataset that aims to fill this gap by combining well-known En- glish resources into one centralized knowledge repository in the form of <term, definition, example> triples. 3D-EX is a unified evalu- ation framework with carefully pre-computed train/validation/test splits to prevent memoriza- tion. We report experimental results that sug- gest that this dataset could be effectively lever- aged in downstream NLP tasks. Code and data are available at https://github.com/ F-Almeman/3D-EX. 1 Introduction Lexicographic definitions have played an impor- tant role in NLP. For example, definitions, and more specifically, term-hypernym pairs occurring in them, constitute a core component in applica- tions such as taxonomy learning (Navigli et al., 2011; Velardi et al., 2013; Espinosa-Anke et al., 2016), knowledge base construction (Delli Bovi et al., 2015), or for augmenting language mod- els (LMs) (Joshi et al., 2020; Chen et al., 2022). For this reason, numerous works have proposed methods to extract definitions from corpora (defini- tion extraction, or DE) (Navigli and Velardi, 2010; Espinosa-Anke and Schockaert, 2018; Spala et al., 2020). However, DE, traditionally framed as a sen- tence classsification problem, plateaus quickly in terms of its applicability to real-world settings for a number of reasons, namely: (1) it is tied to a reference corpus; (2) it does not handle flexible contexts (e.g., definitional information appearing across several sentences); and (3) incorporating monolithic sentence-level definitional knowledge into LMs during pretraining is not straightforward. A complementary task to the above is definition modeling (DM), a promising direction both from resource creation and NLP standpoints. DM is the task of automatically generating human-readable lexicographic definitions or glosses given some in- put. From its inception, where Noraset et al. (2017) trained a bidirectional LSTM on ⟨t, d⟩ pairs, where t is an input term, and d is its corresponding defi- nition, more recent contributions in this area have leveraged contextualized representations by aug- menting t with some context c (Ni and Wang, 2017; Gadetsky et al., 2018; Ishiwatari et al., 2019; Reid et al., 2020; Bevilacqua et al., 2020). A crucial prerequisite for enabling, among oth- ers, successful DM systems is having access to datasets that combine terms, definitions, and good dictionary examples (Kilgarriff et al., 2008; Kosem et al., 2019; …

Content truncated to 500 words